AE on CIFAR10 dataset

The CIFAR-10 dataset consists of 60000 32x32 colour images in 10 classes, with 6000 images per class. There are 50000 training images and 10000 test images.

import data

build AE

encoder

decoder

autoencoder

train AE

save/load model

predict images

categorize by labels/categories

done now

10 classes

1 class

loss for each class

RMSE compared with STD of original image

t-SNE visualization

UMAP visualization

export architecture